CS2125 Paper Review Form - Winter 2019 Reviewer: Yasaman Rohanifar Paper Title: Internet of Vehicles: From Intelligent Grid to Autonomous Cars and Vehicular Clouds Author(s): Mario Gerla, Eun-Kyu Lee, Giovanni Pau, Uichin Lee 1) Is the paper technically correct? [*] Yes [ ] Mostly (minor flaws, but mostly solid) [ ] No 2) Originality [ ] Very good (very novel, trailblazing work) [*] Good [ ] Marginal (very incremental) [ ] Poor (little or nothing that is new) 3) Technical Depth [ ] Very good (comparable to best conference papers) [*] Good (comparable to typical conference papers) [ ] Marginal depth [ ] Little or no depth 4) Impact/Significance [ ] Very significant [*] Significant [ ] Marginal significance. [ ] Little or no significance. 5) Presentation [*] Very well written [ ] Generally well written [ ] Readable [ ] Needs considerable work [ ] Unacceptably bad 6) Overall Rating [ ] Strong accept (award quality) [*] Accept (high quality - would argue for acceptance) [ ] Weak Accept (borderline, but lean towards acceptance) [ ] Weak Reject (not sure why this paper was published) 7) Summary of the paper's main contribution and rationale for your recommendation. (1-2 paragraphs) This paper discusses the evolution of traditional vehicles, to advancements such as intelligent vehicle grid, and leveraging the enormous and ever-growing sensor platform on the road, to the emergence of internet of autnomous vehicles and formation of vehicular cloud. It opens up with the objectives of autonomous vehicles and how the power of existing and growing intelligent sensors has the potential to transform exsiting vehicle fleet to a fully-autonomous fleet by leveraging vehicular clouds. It then discusses some characteristics of the emerging vehicle applications (at the time) that are: Application Content Time-Space Validity, Content Centric Networking, Vehicle Collaboration Sharing Sensory Data, Intelligent Vehicle Grid and Vehicular Data. Later, it introduces the vehicular cloud and discusses a high-level case study of their proposed decentralized network architecure for autonomous vehicles. The last part of the paper is dedicated to studying use cases for vehicular cloud such as Beacons and Alarms, Intelligent Transport, Infrastructure Failure Recovery) and methods and challenges (Virtualization, Security, File and Media Downloading, Cognitive Radios and Spectrum Data-base Crowdsourcing, NDN Network Layer). The authors conclude that the implementation of Internet cloud for vehicles (vehicular cloud) will enable full fleet deployment of autonomous vehicles and will eventually help realize the benefits and promises of AUVs. I believe this paper should be accepted as it presents a novel idea of applying the existing idea of Internet of Things, and leveraging the power of edge computing to realize the dream of full fleet of autonomous vehicles. Though it does not provide high technical aspects of the solutions proposed, it gives an abstract architecture and opens the discourse for future development of such applications in AUVs. 8) List 1-3 strengths of the paper. (1-2 sentences each, identified as S1, S2, S3.) S1: The idea of evolution from internet vehicle grid to autonomous, internet-connected vehicles, and vehicular cloud as well as incorporating edge-computing (in order to take advantage of the road-side infrastructures and existing computing and storage units) was novel at the time that they proposed it. The extension of the idea of Internet of Things to the world of vehicles when full-fleet deployment was far from reality, makes their idea interesting and novel. S2: The paper is very well-written and easy-to-read and it covers most relevant topics for implementation of vehicular cloud and deployment of full-fleet autonomous vehicles based on that. The level of abstraction provided for their envisioned model is comprehensive enough for readers to generally understand how the system works. S3: A good set of use-cases for their envisioned vehicular cloud has been provided. Although it is arguable that some of the important use-cases and challenges such as security have not been elaborated enough. 9) List 1-3 weaknesses of the paper (1-2 sentences each, identified as W1, W2, W3.) W1: Lack of references in my opinion is one of the weaknesses of this paper. Although it can be argued that the issues and challenges they discussed and the future state-of-the-art they envisioned in the paper were not widely available at the time, I still belive that more references could be gathered for this paper. W2: The interesting ideas they discussed needed a little bit more clarification on the details. For example, lots of easy-to-catch challenges in many areas they discussed such as problems with content-centric networking (whether it is secure enough), or the challenges of cloud formation, maintenance, and release were not dicussed. This will lead the audience to make possibly false assumptions about their work or simply be clueless. Furthermore, alternative architectures and paradigms for implementing such systems were not discussed in the paper. W3: The feasibility of implementation of their ideas based on the state-of-the-art at the time were not discussed in the paper. In this regard, this paper can be regarded as a future imagination of autonomous vehicles without taking into account the infrastructure and technical requirements of such vision. Furthermore, some unrealistic assumptions such as the ability of a failed or broken autonomous vehicle to have some level of functionality or decision making were made in this paper that make the feasibility of their envisioned architecture, questionable.